FlowGuard AI Assistant

Security answers, in plain language

An AI security assistant built right into the platform. Employees ask how to stay safe and check their own progress; admins ask about their organisation's risk — answered instantly, grounded only in your own pre-aggregated security intelligence. Never raw data.

Grounded in your data Never sees raw PII Employee & admin Professional & Enterprise

In-App Assistant

Ask. Get grounded answers.

Live
Risk Training Incidents Signals Grounded · never raw data
Two Audiences

One assistant, audience-aware

Everyone gets help in their own words — and only ever sees what their role is allowed to see.

Every Employee

Employee assistant

Self-service answers for the people you're protecting — how to report a suspicious email, where to find their training, and what their own security and risk scores mean. Read-only guidance and navigation, never another person's data.

  • "How do I report a phishing email?"
  • "What's my current training status?"
  • "Is this link safe to click?"
Admins & Security

Admin copilot

Decision support for the people running the programme — organisation-level risk, training completion, and where to focus next. Answers from org aggregates only, never per-employee personal data.

  • "Which departments need attention?"
  • "How is training completion trending?"
  • "How do I launch a phishing campaign?"
How It Works

Helpful, accurate, and private by design

FlowGuard is grounded in your platform's own knowledge and your pre-aggregated security signals — so answers are specific and trustworthy, without the AI ever touching raw customer data.

Grounded in your knowledge

Answers are retrieved from a curated corpus built from the platform's own help, guides, and navigation — the assistant points to real steps and never invents features.

Privacy-first by design

The model only ever receives sanitized, pre-aggregated, redacted context. No raw records, no PII — security correctness is mandatory, not optional.

Audience-scoped retrieval

Employees and admins see different knowledge. An employee can never pull simulation mechanics or another person's data — the boundary is enforced before the model is ever asked.

Answers about your own posture

Ask "what's my risk?" or "what training do I owe?" and get a clear, personalised summary built from minimized, non-identifying facts — no spreadsheets required.

Step-by-step guidance

Instead of a link dump, FlowGuard explains the exact steps to complete a task in the platform — and tells you when to send a help request if it doesn't have the detail.

Always in reach

A floating launcher sits in the topbar and opens a side drawer — help is one click away on any page, for both employees and admins.

Fast and cost-aware

Common questions are handled instantly without calling the model, and responses are cached — so answers are quick and usage stays efficient.

Included on Professional & Enterprise

The FlowGuard Assistant is part of the Professional plan and every plan above it — no add-on, no separate setup.

The AI never sees your raw data

FlowGuard's core principle is simple and absolute: the model only ever receives sanitized, pre-aggregated, redacted context. No raw employee records. No personally identifiable information. Just the minimum needed to give a useful, grounded answer.

It's the AI assistant your security and privacy teams can actually say yes to.

Put a security expert in every employee's pocket

Start your free trial and meet the FlowGuard Assistant — included on Professional and Enterprise.